In this chapter, we describe different methods to estimate an unknown signal from its linear measurements. We focus on the underdetermined case where the number of measurements is less than the dimension of the unknown signal. We introduce the concept of signal sparsity and describe how it could be used as prior information for either regularized least squares or Bayesian signal estimation. We discuss compressed sensing and sparse signal representation as examples where these sparse signal estimation methods could be applied.
Part of the book: Bayesian Inference
We describe two importance sampling techniques for a standard Monte Carlo (MC) method that could enable fast simulation of signals from optical coherence tomography (OCT) imaging systems. These OCT signals are generated due to diffusive reflections from either multilayered or arbitrary shaped, turbid media, for example, tissue. Such signals typically consist of ballistic and quasi-ballistic components, of scattered photons inside the medium, in addition to photons that undergo multiple scattering. We show that MC simulation of these OCT signals using importance sampling reduces its computation time on a serial processor by up to three orders of magnitude compared to its corresponding standard implementation. Therefore, these importance sampling techniques enable practical simulation of OCT B-scans of turbid media, for example, tissue, using commonly available workstations.